arXiv subjects

Sangwon Yoon

Publications and source records attributed to Sangwon Yoon.

8 recordsLinked to original sources

arXiv2026

Automated Hardware Validation Test Plan Generation for Large Scale AI Datacenter Platforms Using a Generative AI Multi-Agents Architecture

Large-scale AI datacenter platforms comprise thousands of heterogeneous hardware components whose validation requires comprehensive fault injection test plans. Today these plans are authored manually: engineers review hardware self-healing validation documents and bills of materials, enumerate failure modes per field-replaceable unit, and produce flat lists of single-layer test cases. This process is labor-intensive, error-prone, and dependent on institutional knowledge; coverage gaps surface late, traceability to source specifications is implicit, and the effort is largely repeated per platform. This paper presents a generative AI multi-agent architecture that automates the generation of structured hardware validation test plans from two canonical inputs: self-healing validation documents, which enumerate known failure modes and their detection and remediation behaviors per field-replaceable unit, and component Bills of Material. An ingestion agent normalizes heterogeneous inputs into a canonical representation; a classification agent maps components to functional domains via contextual reasoning over part descriptions and sub-category hierarchies; and a generation agent synthesizes test cases by combining normalized failure modes with domain-classified data, filling gaps and producing edge cases. The output conforms to a standardized schema for direct import into internal validation software. Evaluated on two production platforms against manual baselines, the framework achieves coverage expansions of 74.2% and 51.4%, cutting authoring from days to hours. It yields fully traceable mappings from each test case to its source specification, and its multi-agent decomposition is portable across platform generations. Automated and expert evaluations confirm 100% extraction fidelity and high acceptance of new scenarios, validating the framework as a robust human-in-the-loop force multiplier.

cs.MA↗

arXiv2024

DEBATE: Devil's Advocate-Based Assessment and Text Evaluation

As natural language generation (NLG) models have become prevalent, systematically assessing the quality of machine-generated texts has become increasingly important. Recent studies introduce LLM-based evaluators that operate as reference-free metrics, demonstrating their capability to adeptly handle novel tasks. However, these models generally rely on a single-agent approach, which, we argue, introduces an inherent limit to their performance. This is because there exist biases in LLM agent's responses, including preferences for certain text structure or content. In this work, we propose DEBATE, an NLG evaluation framework based on multi-agent scoring system augmented with a concept of Devil's Advocate. Within the framework, one agent is instructed to criticize other agents' arguments, potentially resolving the bias in LLM agent's answers. DEBATE substantially outperforms the previous state-of-the-art methods in two meta-evaluation benchmarks in NLG evaluation, SummEval and TopicalChat. We also show that the extensiveness of debates among agents and the persona of an agent can influence the performance of evaluators.

cs.CL↗

arXiv2023

Corporate Bankruptcy Prediction with Domain-Adapted BERT

This study performs BERT-based analysis, which is a representative contextualized language model, on corporate disclosure data to predict impending bankruptcies. Prior literature on bankruptcy prediction mainly focuses on developing more sophisticated prediction methodologies with financial variables. However, in our study, we focus on improving the quality of input dataset. Specifically, we employ BERT model to perform sentiment analysis on MD&A disclosures. We show that BERT outperforms dictionary-based predictions and Word2Vec-based predictions in terms of adjusted R-square in logistic regression, k-nearest neighbor (kNN-5), and linear kernel support vector machine (SVM). Further, instead of pre-training the BERT model from scratch, we apply self-learning with confidence-based filtering to corporate disclosure data (10-K). We achieve the accuracy rate of 91.56% and demonstrate that the domain adaptation procedure brings a significant improvement in prediction accuracy.

cs.CL↗

arXiv2023

Detecting Rumor Veracity with Only Textual Information by Double-Channel Structure

Kyle (1985) proposes two types of rumors: informed rumors which are based on some private information and uninformed rumors which are not based on any information (i.e. bluffing). Also, prior studies find that when people have credible source of information, they are likely to use a more confident textual tone in their spreading of rumors. Motivated by these theoretical findings, we propose a double-channel structure to determine the ex-ante veracity of rumors on social media. Our ultimate goal is to classify each rumor into true, false, or unverifiable category. We first assign each text into either certain (informed rumor) or uncertain (uninformed rumor) category. Then, we apply lie detection algorithm to informed rumors and thread-reply agreement detection algorithm to uninformed rumors. Using the dataset of SemEval 2019 Task 7, which requires ex-ante threefold classification (true, false, or unverifiable) of social media rumors, our model yields a macro-F1 score of 0.4027, outperforming all the baseline models and the second-place winner (Gorrell et al., 2019). Furthermore, we empirically validate that the double-channel structure outperforms single-channel structures which use either lie detection or agreement detection algorithm to all posts.

cs.CL↗

arXiv2023

Supersymmetric extension of universal enveloping vertex algebras

In this paper, we study the construction of the supersymmetric extensions of vertex algebras. In particular, for $N = n \in \mathbb{Z}_{+}$, we show the universal enveloping $N = n$ supersymmetric (SUSY) vertex algebra of an $N = n$ SUSY Lie conformal algebra can be extended to an $N = n' > n$ SUSY vertex algebra.

math-ph↗

arXiv2023

Superconformal structures of supersymmetric free fermion vertex algebras

In this paper, we define the shifted superconformal vector of supersymmetric charged free fermion vertex algebras, which is a 1-parameter deformation of the superconformal vector of the SUSY $bc$-$\beta\gamma$ system. Moreover, we find the corresponding shifted $N=2$ superconformal symmetry of SUSY charged free fermion vertex algebras, by using the $N_{K}=1$ SUSY vertex algebra formalism. Finally, in order to describe the shifted $N=2$ superconformal symmetry of the SUSY charged free fermion vertex algebra by $N=2$ superfields, we construct an $N_{K}=2$ SUSY version of the $bc$-$\beta\gamma$ system.

math-ph↗

arXiv2022

Development of a Scanning Tunneling Microscope for Variable Temperature Electron Spin Resonance

Recent advances in increasing the spectroscopic energy resolution in scanning tunneling microscopy (STM) have been achieved by integrating electron spin resonance (ESR) with STM. Here, we demonstrate the design and performance of a home-built STM capable of ESR at temperatures ranging from 1 K to 10 K. The STM is incorporated with a home-built Joule-Thomson refrigerator and a 2-axis vector magnet. Our STM design allows for the deposition of atoms and molecules directly into the cold STM, eliminating the need to extract the sample for deposition. In addition, we adopt two methods to apply radio-frequency (RF) voltages to the tunnel junction, the early design of wiring to the STM tip directly, and a more recent idea to use an RF antenna. Direct comparisons of ESR results measured using the two methods and simulations of electric field distribution around the tunnel junction show that, despite their different designs and capacitive couplings to the tunnel junction, there is no discernible difference in the driving and detection of ESR. Furthermore, at a magnetic field of 1.6 T, we observe ESR signals (near 40 GHz) sustained up to 10 K, which is the highest temperature for ESR-STM measurement reported to date, to the best of our knowledge. Although the ESR intensity exponentially decreases with increasing temperature, our ESR-STM system with low noise at the tunnel junction allows us to measure weak ESR signals with intensities in the sub-fA range. Our new design of ESR-STM, which is operational in a large frequency and temperature range, can broaden the use of ESR spectroscopy in STM and enable the simple modification of existing STM systems, which will hopefully accelerate a generalized use of ESR-STM.

cond-mat.mes-hall↗

arXiv2021

Spin Resonance Amplitude and Frequency of a Single Atom on a Surface in a Vector Magnetic Field

We used electron spin resonance (ESR) combined with scanning tunneling microscopy (STM) to measure hydrogenated Ti (spin-1/2) atoms at low-symmetry binding sites on MgO in vector magnetic fields. We found strongly anisotropic g-values in all three spatial directions. Interestingly, the amplitude and lineshape of the ESR signals are also strongly dependent on the angle of the field. We conclude that the Ti spin is aligned along the magnetic field, while the tip spin follows its strong magnetic anisotropy. Our results show the interplay between the tip and surface spins in determining the ESR signals and highlight the precision of ESR-STM to identify the single atom's spin states.

cond-mat.mes-hall↗